The combustion behaviour of ammonia has attracted intermittent interest with the original patent by
Lyon (US3900554A) relating to its use for nitric oxide reduction through selective non-catalytic reduction
(SNCR) a focal point. The recent interest in ammonia stems from its use as a hydrogen rich energy carrier
with practical use requiring a much wider parameter space. The corresponding challenges (e.g. Kobayashi
et al., Proc. Combust. Inst. 37 (2019) 109–133) include different flame dynamics and high emissions of
oxides of nitrogen. The current paper explores the complex nature of ammonia oxidation and provides
a reduced size reaction mechanism that enables application, without approximation, to the computation
of turbulent flames through a joint-scalar transported probability density function (JPDF) method. Comprehensive validation suggests similar accuracy to a reference mechanism (Glarborg et al., Prog. Energy
Combust. Sci. 67 (2018) 31–68) and highlights some uncertainties. The selected turbulent flame configuration features auto-ignition stabilised flames supported by a coflow of hot combustion products. The
base case features a H2/N2 fuel jet that permits flame stabilisation at 1045 K corresponding to the onset of the SNCR temperature window. The impact of a gradual substitution of hydrogen by ammonia on
flame stabilisation, emissions of oxides of nitrogen and the flame structure is quantified. It is shown that
ammonia substitution leads to more prevalent local extinction, a more distributed flame structure and
requires substantially increased coflow temperatures to achieve a similar flame stabilisation point. A lowering of the coflow temperature to operate within the SNCR regime substantially reduces NOx and leads
towards a homogeneous/distributed reaction mode. The reduced fuel reactivity highlights the importance
of turbulence-chemistry interactions leading to complexities in the design of practical devices.